Anyone using "Amazon Marketing Services" for Books?

[Justin Gesso]

I think the title says it.

Amazon has its own advertising platform, which looks like an immature version of Facebook's and Google's advertising platforms.

Basically, you pay per click, and your book shows in the "Sponsored Products" or various other locations on Amazon.com.

Is anyone here using it successfully to drive sales to their books?

I have it set up and am getting a decent return (ad cost is averaging lifetime 26% of sales volume, which leaves room for profit), but it's tough to optimize due to limitations in the interface and bare-bones documentation.

Some questions I have...

1. Basic freakin' UI question...is there a way to set the date range for the data you're looking at?
2. Are there any best practices for determining keywords and bid amounts (apart from burning your own cash with experiments)?
3. Any recommendations on ad placement types (product display, sponsored products, etc...)?
4. Is there a good benchmark for ad cost/sales volume?

Thanks!

PS - Apologies if this has already been covered. I did a forum search, but didn't see this specific topic.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[amp0193]

Basic freakin' UI question...is there a way to set the date range for the data you're looking at?

No. You have to go to "reports" and set the date range, and then you download a spreadsheet.

Are there any best practices for determining keywords and bid amounts (apart from burning your own cash with experiments)?

If you're concerned about cash, stick to keywords that are directly relevant to your product/book. However, there's going to be some experimenting no matter what. The default mode is "phrase match" which is going to pull in customer search terms that include that phrase. Exact match will only show searches for that keyword. You can use a site like textmechanic, or search for a keyword generator to mix up the words to create many different combinations of possible searches. Know what your break-even ACOS is, and when keywords are pushing beyond that lower the bid or drop the keyword.

Any recommendations on ad placement types (product display, sponsored products, etc...)?

Sponsored products is no different than running sponsored products through Seller Central, except the reporting is way worse. Works great, if you're not already doing it through SC. Headline ads kill it. Make a good headline, and have it drop on a "brand page" and you'll do really well with these. The CPC is significantly lower with headline ads than with sponsored products. Typically half as much cost in mine and others experience.

Is there a good benchmark for ad cost/sales volume?

Totally dependent on niche and competition. Do the math, and know when you are losing money, and don't let that happen. In general, I would say that under 20% ACoS is good, and that under 10% is great.

 

[amp0193]

Also, we got into AMS quite a bit in this thread: https://www.thefastlaneforum.com/co...e-vastly-different-than-amazon-sellers.65526/

Ignore the drama.

 

[Justin Gesso]

No. You have to go to "reports" and set the date range, and then you download a spreadsheet.



If you're concerned about cash, stick to keywords that are directly relevant to your product/book. However, there's going to be some experimenting no matter what. The default mode is "phrase match" which is going to pull in customer search terms that include that phrase. Exact match will only show searches for that keyword. You can use a site like textmechanic, or search for a keyword generator to mix up the words to create many different combinations of possible searches. Know what your break-even ACOS is, and when keywords are pushing beyond that lower the bid or drop the keyword.



Sponsored products is no different than running sponsored products through Seller Central, except the reporting is way worse. Works great, if you're not already doing it through SC. Headline ads kill it. Make a good headline, and have it drop on a "brand page" and you'll do really well with these. The CPC is significantly lower with headline ads than with sponsored products. Typically half as much cost in mine and others experience.




Totally dependent on niche and competition. Do the math, and know when you are losing money, and don't let that happen. In general, I would say that under 20% ACoS is good, and that under 10% is great.

Beautiful. Thank you very much. Such great info in your responses, @amp0193.

I will also review that thread and am well versed in ignoring the drama. ;-)

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[amp0193]

Beautiful. Thank you very much. Such great info in your responses, @amp0193.

I will also review that thread and am well versed in ignoring the drama. ;-)

I should also mention, that product display has been a complete bust for me, and the same for someone else I know who's tried it. But you never know for yourself until you test. PPC is all about testing. Hell, everything is all about testing (Speaking of, if you aren't using Split.ly, you should be).

Suggested best practices for Product Display (from a AMS rep) are:

Run a campaign targeting all of your ASINs - "protect your territory"

Run a campaign targeting competitor ASINs. Can work well if you product is similar, but at a lower cost than the competitor.

Run a campaign targeting an "interest group". This is so widespread... I can't imagine that a book would do well with interest targeting.

Run a campaign targeting similar products that "pair" well with yours.

Product Display will get you 120,000 displays for like $10... but maybe no sales either. I have yet to ever get a sale from my product display campaigns. Maybe in the long run it's worth the brand recognition for the huge number of displays... but I like tangible results.

 

[Justin Gesso]

I should also mention, that product display has been a complete bust for me, and the same for someone else I know who's tried it. But you never know for yourself until you test. PPC is all about testing. Hell, everything is all about testing (Speaking of, if you aren't using Split.ly, you should be).

Suggested best practices for Product Display (from a AMS rep) are:

Run a campaign targeting all of your ASINs - "protect your territory"

Run a campaign targeting competitor ASINs. Can work well if you product is similar, but at a lower cost than the competitor.

Run a campaign targeting an "interest group". This is so widespread... I can't imagine that a book would do well with interest targeting.

Run a campaign targeting similar products that "pair" well with yours.

Product Display will get you 120,000 displays for like $10... but maybe no sales either. I have yet to ever get a sale from my product display campaigns. Maybe in the long run it's worth the brand recognition for the huge number of displays... but I like tangible results.

Really good information again. Thank you.

I have not come across split.ly before, but looks like it fills in the gaps I have here.

Good to know about Product Display.

The "Sponsored Products" option seems to be working well for me now. I think I'll start by drilling into that more with keyword and bid experiments.

Again, thanks so much.

 

[amp0193]

The "Sponsored Products" option seems to be working well for me now. I think I'll start by drilling into that more with keyword and bid experiments.

You can run multiple campaigns. You can have a sponsored products campaign and a headline campaign simultaneously. Headline campaigns are half as cheap... try it out. You can take the good performing keywords from Sponsored products, and use those, if you want to do it that way.

For my main keyword, I have a big headline banner show up, as well as an in-line Sponsored Products ad at the top of search, as well as 1 of the right side sponsored product ad boxes for a different product. And then I have 5-6 products on the first page orgnaically.

It never hurts to beat the audience over the head to get them to notice you.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Justin Gesso]

It never hurts to beat the audience over the head to get them to notice you.

Very true.

 

[AllenCrawley]

I have yet to ever get a sale from my product display campaigns. Maybe in the long run it's worth the brand recognition for the huge number of displays... but I like tangible results.

Product display ads do pretty well for us. ACoS is around 30%.

 

[Justin Gesso]

Product display ads do pretty well for us. ACoS is around 30%.

Oh really? I will give them another shot. Worth the experiment.

I'm getting an ACoS of 18.91% on the Sponsored, but the ads are not getting much exposure and nowhere near my spending limit. I've tried various bid levels and continue adding/experimenting with new keywords.

I'll start some Product Display ones today.

Thanks for the info!

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[amp0193]

Product display ads do pretty well for us. ACoS is around 30%.

Since I made that post, I had started some new product display campaigns.

The ones targeting my competitors did horribly... 110% ACoS

The ones targeting my own products did better ~40% ACoS.


Today, I reduced bid super low on the competitor one... maybe I will pull some random sales on lower traffic ASINs.

I also reduced the bid a bit on my own products to get that ACoS under 30%.

 

[Justin Gesso]

Since I made that post, I had started some new product display campaigns.

The ones targeting my competitors did horribly... 110% ACoS

The ones targeting my own products did better ~40% ACoS.


Today, I reduced bid super low on the competitor one... maybe I will pull some random sales on lower traffic ASINs.

I also reduced the bid a bit on my own products to get that ACoS under 30%.

Cool, I'd be interested in the results when they roll in.

 

[amp0193]

Cool, I'd be interested in the results when they roll in.

Sure, I'll share.

The most frustrating thing, is that you can't see a breakdown of clicks/traffic per ASIN that you're targeting. Otherwise, I would be optimizing and removing non-profitable ASINs. The only option is to reduce overall bid, unfortunately.

I guess you could test, by making 20-100 campaigns, each targeting 1 competitor ASIN, and then keep the profitable ones and put them all into one campaign that you can manage... but that sounds like a huge PIA, for what would be a small return.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Justin Gesso]

Sure, I'll share.

The most frustrating thing, is that you can't see a breakdown of clicks/traffic per ASIN that you're targeting. Otherwise, I would be optimizing and removing non-profitable ASINs. The only option is to reduce overall bid, unfortunately.

I guess you could test, by making 20-100 campaigns, each targeting 1 competitor ASIN, and then keep the profitable ones and put them all into one campaign that you can manage... but that sounds like a huge PIA, for what would be a small return.

Yes, the entire UI is very basic. Very limited ability to break down the data. I hope to see the platform mature. Just a freakin basic date range box would be nice.

I agree...for now, it just seems you have to have a ton of campaigns running.

 

[amp0193]

Just a freakin basic date range box would be nice.

Yes, I was optimizing some campaigns today and am very frustrated by the lack of ability to filter by date. If drop bid a bit on an unprofitable keyword that is making sales... there's not a way to see the results of that change.

Maybe the best process is to create an "updated keywords campaign", where you put keywords with new bids, and then remove them from their previous campaign.

I'll do that next time I optimize.

 

[MJ DeMarco]

Any updates? This feature is relatively new for books -- curious to how it is going.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Justin Gesso]

Any updates? This feature is relatively new for books -- curious to how it is going.

Yes...

The Good

• In short, it is worthwhile.
• The ROI is there. I spend less in advertising than it takes to cover the margins on the books sold.
• It maintains my best seller positions. I just turned advertising off this week to see what would happen, and my book immediately plunged from positions 1, 2, or 3 in several categories. I still show as top 10 in some sub-categories, but it's a very noticeable effect. Reducing price has not countered it. I plan on raising price back up and kicking advertising back on.
• Sponsored Products ad types work for me. I've run 5 campaigns, all have good ROI.

The Bad

• As you can see in the comments in this thread, the platform is very immature. It makes it super difficult to apply "scientific advertising" principles and test thoroughly. The basics a marketer needs are just missing.
• It won't spend my money! I have an awesome ROI so want to throw more money at this. I currently have a $75 per day limit, but it's spending nowhere near that. Google and FB always seem to figure out how to spend my money.
• I've tried experimenting with different bid levels and keywords. I've done plenty of research, but I really just can't seem to get it to move faster. This is made more difficult by the fact that the analytics just aren't there.
• Product Display ads do not work for me. I've run 2 campaigns and both have negative ROI.

So if you're an author...and not competing with me...do it! And share best practices, findings, and results here please.

 

[amp0193]

• Sponsored Products ad types work for me. I've run 5 campaigns, all have good ROI.

• As you can see in the comments in this thread, the platform is very immature. It makes it super difficult to apply "scientific advertising" principles and test thoroughly. The basics a marketer needs are just missing.

You'll get much more advanced reporting by running your Sponsored Products ads through Seller Central. The cost for Sponsored Products through AMS vs. through Seller Central weren't any different for me.

 

[Justin Gesso]

You'll get much more advanced reporting by running your Sponsored Products ads through Seller Central. The cost for Sponsored Products through AMS vs. through Seller Central weren't any different for me.

This has been somewhat of a confusing topic. Since I'm using KDP and Createspace (self-publishing), I don't have an Amazon Seller Central account.

I'm going off this fairly complex guide to where authors conduct Amazon business: https://authorcentral.amazon.com/gp/help?ie=UTF8&topicID=200650270

Through KDP, I'm given a limited ad dashboard. Sounds like it's more robust on the Seller Central side.

This is a whole separate can of worms, but perhaps an alternate printing/fulfillment option would give me better advertising.

As it stands, I decided to try layering in other advertising methods through FB. I'm using Amazon Associates Tracking IDs to figure out which offer solid ROI.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[amp0193]

This has been somewhat of a confusing topic. Since I'm using KDP and Createspace (self-publishing), I don't have an Amazon Seller Central account.

I'm going off this fairly complex guide to where authors conduct Amazon business: https://authorcentral.amazon.com/gp/help?ie=UTF8&topicID=200650270

Through KDP, I'm given a limited ad dashboard. Sounds like it's more robust on the Seller Central side.

Ahhhh, ok. I've literally never heard of KDP. Didn't know that was a thing. Carry on

 

[AlphaWulf]

I didn't even know Amazon had anything like this. Thanks for giving me an idea!

 

[3things]

I've used this quite a lot at work (run online advertising for large publisher), when it was in beta-only...and I'd echo most of the findings above. My observations to the rep's I've spoken to (we were required to give feedback) have all been along the lines of:

1. Lack of volume (I struggled to spend $300 of a $5k budget in about 3 weeks - you end up going broader to get some volume, but of course that kills the ROI)
2. Reporting was pretty useless for us (it'd be fine if you self pub, and have idk, 1-10 books) - we have 30,000 titles and their 'conversion' is a sale of any of those books that happen in a 2 week window if the person saw your ad - so a pretty vague 'view through' type cookie, that resulted in some pretty misleading +ROI stats. They have since amended this (I'm guessing I wasn't the only one who complained about this..) and I got an email from the Amazon rep the other day saying (from memory) product/ASIN level sales reporting was now being given. Think you have to download a report to see this, it's not displayed on-screen.
3. The +ROI is great (getting this on a $3-$9 fiction book is pretty damn impossible!)...but its small relative to effort (not moving needles for us), and it was already disappearing as more publishers came into the beta, bids went up even on generic terms as we all chased volume on more generic previously profitable terms. As soon as I started seeing -ROI, for me its just yet another ad platform as the USP had gone, plus lacking any volume.
4. Did i mention it seemed to lack volume?

My opinion overall of the platform was that it would suit smaller self-pub authors really nicely, but not large publishers like us (lack of scale, opaque reporting at the time) - the ROI is fine and dandy, but managing this, setting it all up, optimising it...is pretty time consuming relative to the scale that seems available...especially when you figure that as more people use it, the ROI is going to head south anyway.

Not to derail, but I'm a bigger fan of the BookBub self-serve ads that they serve into their emails, have seen consistent success with those at work and they're super fast to set up. You're better off with similar author (rather than category) targeting just like on FB, which increases relevance and therefore CTR, but will reduce volume. I found the best way for BookBub was to run in 1-2 day bursts, spending about $500/day. Letting it run longer just burned out the CTR as you're getting the same people day after day. Worth looking into.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Deleted5250X]

Thanks for this thread guys, it is always interesting when I read about how people are marketing and how they're managing those campaigns.

However, there were a few terms I didn't recognize, so I looked up the definitions and am posting them for others like me.

ACoS or ACOS - Advertising Cost of Sale, this is the percent of sales spent on Advertising. If a product costs $4 and $1 is spent on advertising the ACoS is 25%

ASIN - Amazon Standard Identification Number, a unique identification number identifying a product on the Amazon Marketplace. For books this is the same as the ISBN.

Sources:

https://www.amazon.com/gp/help/customer/display.html?ie=UTF8&nodeId=201563200

https://en.m.wikipedia.org/wiki/Amazon_Standard_Identification_Number